Dendritic cell (DC)-based immunotherapy is a promising approach for cancer treatment. Phase I clinical trials employing adoptive transfer of cultured DCs pulsed with tumor antigens ex vivo have recently been initiated. Genetic transduction of DCs represents another method of tumor antigen presentation with potential advantages over pulsing. Although several such genetic vaccine strategies show promise, optimal vectors and delivery methods are still being developed. The present proposal examines the use of recombinant adeno-associated virus (rAAV) vectors for ex vivo transduction of DCs, for the purpose of anti-tumor immunization. Potential advantages of rAAV for this application include limited host exposure to viral structural elements and stable expression of the antigen. Preliminary data indicate that transduction of ex vivo differentiated human and mouse DCs by rAAV is possible. Further, human peripheral blood monocytes transduced with rAAV and subsequently differentiated, showed robust transgene expression in the generated DCs. Based on our previous studies that transduction of recombinant adenoviral (Ad) vectors through alternate cellular pathways in DCs promote both DC-specific targeting and DC-maturation, we hypothesize that retargeting AAV vectors through the CD40 pathway will similarly promote DC-specific targeting and activation. The goals of this project are to optimize DC transduction by rAAV encoding a tumor antigen, carcinoembryonic antigen (CEA), to achieve potent T cell responses and to evaluate the efficacy of rAAV-CEA transduced DCs in antitumor activity in a mouse model in vivo. The proposed experiments will provide a necessary proof of principle in support of future trials of rAAV-CEA transduced DC vaccines for immunotherapy of CEA-positive human cancers and also will have relevance for the development of additional rAAV vaccines expressing other antigens or genes to enhance antigen-presenting functions.